Method and system for network fault management

Abstract

Methods of processing record data are presented, together with a Central Management Node for processing record data, and Management Subsystem DataBases for providing a record data set and matching record data. Triggered by service related network fault event, a key table comprising at least one key type is determined, based on a focus target record and related to the network fault event is determined, which key table is used in the step matching the at least one key with record data sets in determined Management Subsystem Databases. Matched results are merged, whereafter a service performance indicator may be determined, based on which a multi-layer water flow analysis of subsystem specific performance related indicators can be obtained. This analysis may be successfully used to reveal the location of the service related network fault.

Description

TECHNICAL FIELDThe present invention relates in general to methods and apparatuses for network fault management and in particular to methods and apparatuses for progressive correlation enabling localization of network faults.BACKGROUNDKnown fault localization methods aim at finding the correlation between fault carrying events and a network fault. This is usually a difficult task due to the relatively large amount of events caused by a network fault.An important limitation of such methods is that they limit the scope of correlation to fault carrying events. In reality, in a complex communication system, such as mobile systems, it may not be necessarily sufficient to process fault events alone to succeed in fault localization, because fault events themselves may be an effect of a real fault and may even occur as symptoms of the real fault. These symptoms may be localized far away from the actual location of the fault.For example, subsystem I has a retransmit timer, which can be set s...

AI Technical Summary

Benefits of technology

The present invention provides an improved method for localizing network faults associated with a service. The method involves obtaining information related to the service-related network fault, determining a focus target record, selecting at least one peripheral database, and determining a key table comprising at least one key type related to the record data set. The method also involves matching the key type of the key table with a database record set, attaching matched record data to the key table, and merging the matched record data with the obtained record data set. This allows for a data record information set that includes the matched database records and the determined at least one record set. The invention also provides a network fault management system that includes a central management node and a management subsystem database for processing and matching record data sets for enabling localization of network faults. The system can also provide a valid end-to-end performance survey and allow for self-tuning methods.

Problems solved by technology

This is usually a difficult task due to the relatively large amount of events caused by a network fault.

An important limitation of such methods is that they limit the scope of correlation to fault carrying events.

In reality, in a complex communication system, such as mobile systems, it may not be necessarily sufficient to process fault events alone to succeed in fault localization, because fault events themselves may be an effect of a real fault and may even occur as symptoms of the real fault.

For example, subsystem I has a retransmit timer, which can be set so that a service II experience a large access delay because of that its state machine delay requirements are often exceeded, which now is due to the interaction between the retransmit timer of subsystem I and its own state machine.

Moreover, common communication systems have in fact limited the set of possible fault events, for the reason that it may be difficult to represent all possible problem sources.

For example, routers typically report lost packets.

However, they do not report packet reordering.

Nevertheless, there are services that are sensitive to high levels of reordering of packets.

However, these techniques seem to search for dependencies between the fault event(s) and the problem causing the fault events, within one or more different subsystems.

However these are not applicable to system levels, for the reason that they fail to relate the various services to different subsystems.

From US-2003018228-A1 it is known a graph-based dependence mapping technique, which describes fixed dependences and which is therefore usually not applicable since this assumption is generally not feasible.

Images